INDEX

Explanations

words related to locations and activities associated with various professions or hobbies

oai_token-act-pair · gpt-3.5-turbo

references to subjects and their actions or characteristics within various contexts

oai_token-act-pair · gpt-4o-mini Triggered by @bot

New Auto-Interp

Top Features by Cosine Similarity

Comparing With GPT2-SMALL @ 4-res-jb

Configuration

jbloom/GPT2-Small-SAEs-Reformatted/blocks.4.hook_resid_pre

Prompts (Dashboard)

24,576 prompts, 128 tokens each

Dataset (Dashboard)

Skylion007/openwebtext

Features

24,576

Data Type

torch.float32

Hook Point

blocks.4.hook_resid_pre

Architecture

standard

Context Size

128

Dataset

Skylion007/openwebtext

Hook Point Layer

Activation Function

relu

Embeds

IFrame

Link

Not in Any Lists

No Comments

Negative Logits

fty

-0.77

sav

-0.71

pps

-0.69

cause

-0.69

anova

-0.67

henko

-0.65

 Saying

-0.64

foundland

-0.63

csv

-0.62

etheless

-0.62

POSITIVE LOGITS

 intersect

0.88

 reside

0.78

 resided

0.77

 supposedly

0.76

 congreg

0.76

 resides

0.73

 awaited

0.72

 counted

0.70

 surrounded

0.70

 deemed

0.68

Activations Density 0.485%

words related to locations and activities associated with various professions or hobbies

references to subjects and their actions or characteristics within various contexts

No Comments

No Known Activations